Institutional Repository of Key Laboratory of Behavioral Science, CAS
Depressive Emotion Recognition Based on Behavioral Data | |
Yue Su1,2; Huijia Zheng1,3; Xiaoqian Liu1; Tingshao Zhu1 | |
2019 | |
通讯作者邮箱 | tszhu@psych.ac.cn |
会议名称 | Human-Centered Computing (HCC 2018) |
会议录名称 | HCC 2018 |
会议日期 | 2018.12 |
会议地点 | 墨西哥 |
摘要 |
With the increase of pressure in people’s lives, depression has become one of the most common mental illness worldwide. The wide use of social media provides a new platform for depression recognition based on people’s behavioral data. This study utilizes the linguistical psychological characteristics of Weibo users to predict users’ depression level. The model adopts the Gaussian process regression algorithm, sets the PUK kernel as the kernel function, applies the forward-backward search method to select feature, and uses five-fold cross-validation to evaluate performance of the model. This study finally established a prediction model with a correlation coefficient of 0.5189, which achieved a medium correlation in the psychological definition, and provided a more accurate method for the auxiliary diagnosis of depression. |
关键词 | Diseases Learning systems Models Social networking |
收录类别 | EI |
语种 | 英语 |
文献类型 | 会议论文 |
条目标识符 | http://ir.psych.ac.cn/handle/311026/29137 |
专题 | 中国科学院行为科学重点实验室 |
通讯作者 | Tingshao Zhu |
作者单位 | 1.Institute of Psychology, Chinese Academy of Sciences 2.Department of Psychology, University of Chinese Academy of Sciences 3.St. Mark’s School, Southborough |
通讯作者单位 | 中国科学院心理研究所 |
推荐引用方式 GB/T 7714 | Yue Su,Huijia Zheng,Xiaoqian Liu,et al. Depressive Emotion Recognition Based on Behavioral Data[C],2019. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Depressive Emotion R(10747KB) | 会议论文 | 限制开放 | CC BY-NC-SA | 请求全文 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论